I bought a FujiFilm X-E5 recently, which is an amazing camera for size and money, but I shocked at how miserable their iOS app is. It takes several minutes per photo to transfer. The camera and phone run out of battery before a small memory card is transferred.
So I got Claude to help me reverse engineer their protocol using a mixture of BT and Wifi capture tools, then built a native app in Swift that works on Mac and iOS and by disabling a bunch of on camera processing I got it to transfer a photo in about 3 seconds over wifi bridge. I’ve been able to multithread this to 3 concurrent downloads without crashing the camera.
I did, they asked me to open a support ticket, which I did, and the last response I got was:
> We've reviewed the details you provided and updated your case with the necessary information. It is now being routed to the appropriate team for further support.
Has anyone here used Telnyx? I tried to build a product against their API last year and 3 weeks after signing up they banned my account and made it impossible to get an answer as to why or re-enable it.
The last time I applied for a credit card (about 4 years ago) in Australia, the bank used an app that read the photo page and chip of my passport to verify that it was a real document. That process does verify the authenticity of the document.
I’ve been playing around with Zed, supports local and cloud models, really fast, nice UX. It does lack some of the deeper features of VSCode/Cursor but very capable.
Service account file vs API Key have similar security risks if provided the way you are using them. Google recommends using ADC and it’s actually an org policy recommendation to disable SA files.
This happened to me in Bali a few months ago, I ordered a plate of tacos, they came with a side of lime. When I squirted the lime some of it shot out the side and got me right in the forehead. I didn’t think anything of it until it became inflamed and later blistered. I still have a dark patch resembling a drip on my forehead.
I’m curious to hear more about your pivot from AI workflow builder to document parsing. I can see correlations there, but that original idea seems like a much larger opportunity than parsing PDFs to tables in what is an already very crowded space. What verticals did you find have this problem specifically that gave you enough conviction to pivot?
I was working on a project where GitHub copilot refused to work in a currencies.go file because it contained “Vietnamese Dong”. Removing that line made it work.